Estimation of general nonlinear adsorption isotherms from chromatograms

Adsorption isotherms are the most important parameters in rigorous models of chromatographic processes. In this paper, a numerical method to estimate them from chromatograms is proposed. In contrast to existing numerical methods, the isotherms are represented by neural networks. The weights and the biases of neural networks are optimized by a nonlinear least squares algorithm. Due to the universal approximating capability of neural networks, this method could theoretically retrieve any form of isotherm from the chromatograms. Several issues of the method are studied, which include the chosen type of neural networks, the choice of the mapping relationship, the initialization of neural networks, the importance of the experimental design used to generate the chromatograms and the consideration of the physical role of the isotherms. The potential of the new method is illustrated by the application to the isotherm estimation for EMD53986 on a chiral phase from measured batch elution chromatograms.

[1]  Sebastian Engell,et al.  Novel Learning Tasks, Optimization, and Their Application , 2003 .

[2]  Mauricio Sepúlveda,et al.  Parameter identification for a model of chromatographic column , 1994 .

[3]  Shang-Liang Chen,et al.  Orthogonal least squares learning algorithm for radial basis function networks , 1991, IEEE Trans. Neural Networks.

[4]  Anita M. Katti,et al.  Fundamentals of Preparative and Nonlinear Chromatography , 1994 .

[5]  Douglas M. Ruthven,et al.  Principles of Adsorption and Adsorption Processes , 1984 .

[6]  Anita M. Katti,et al.  Simple wave effects in two-component nonlinear liquid chromatography: application to the measurement of competitive adsorption isotherms , 1990 .

[7]  P. Kipping,et al.  Measurement of Adsorption Isotherms by a Gas Chromatographic Technique , 1965, Nature.

[8]  E. Glueckauf,et al.  Theory of chromatography. Part 10.—Formulæ for diffusion into spheres and their application to chromatography , 1955 .

[9]  T Poggio,et al.  Regularization Algorithms for Learning That Are Equivalent to Multilayer Networks , 1990, Science.

[10]  Sebastian Engell,et al.  Neural Network-Based Identification of Nonlinear Adsorption Isotherms , 2004 .

[11]  E. Glueckauf,et al.  239. Theory of chromatography. Part II. Chromatograms of a single solute , 1947 .

[12]  Mauricio Sepúlveda,et al.  Determination of binary competitive equilibrium isotherms from the individual chromatographic band profiles , 1999 .

[13]  C. S. G. Phillips,et al.  The chromatography of gases and vapours. Part III. The determination of adsorption isotherms , 1954 .

[14]  F. Helfferich,et al.  Accurate Chromatographic Method for Sorption Isotherms and Phase Equilibria , 1963, Science.

[15]  Erika Cremer,et al.  Messung von Adsorptionsisothermen an Katalysatoren bei hohen Temperaturen mit Hilfe der Gas-Festkörper-Eluierungschromatographie , 1961 .

[16]  Igor Quiñones,et al.  Application of different isotherm models to the description of single-component and competitive adsorption data , 1996 .

[17]  Hans-Paul Schwefel,et al.  Advances in Computational Intelligence: Theory and Practice , 2002 .

[18]  Tingyue Gu,et al.  Mathematical Modeling and Scale-up of Liquid Chromatography , 1995 .

[19]  Kurt Hornik,et al.  Multilayer feedforward networks are universal approximators , 1989, Neural Networks.

[20]  Stephen C. Jacobson,et al.  Determination of isotherms from chromatographic peak shapes , 1991 .

[21]  Georges Guiochon,et al.  Numerical determination of the competitive isotherm of enantiomers. , 2003, Journal of chromatography. A.